The Potential Role of Carbon Labeling in a Green Economy

Energy Economics 34 (2012) S53–S63

Contents lists available at SciVerse ScienceDirect

Energy Economics

journal homepage: www.elsevier.com/locate/eneco

The potential role of carbon labeling in a green economy

Mark A. Cohen a,b,⁎, Michael P. Vandenbergh c a Vanderbilt University, United States b Resources for the Future, United States c Climate Change Research Network, Vanderbilt Law School, United States article info abstract

Article history: Over the past several years, labeling schemes that focus on a wide range of environmental and social metrics Received 30 January 2012 have proliferated. Although little empirical evidence has been generated yet with respect to carbon footprint Received in revised form 22 May 2012 labels, much can be learned from our experience with similar product labels. We first review the theory and Accepted 30 August 2012 evidence on the role of product labeling in affecting consumer and firm behavior. Next, we consider the role Available online 7 September 2012 of governments and nongovernmental organizations, concluding that international, multistakeholder organizations have a critical part to play in setting protocols and standards. We argue that it is important to consider JEL classification: D82 the entire life cycle of a product being labeled and develop an international standard for measurement and F18 reporting. Finally, we examine the potential impact of carbon product labeling, discussing methodological K32 and trade challenges and proposing a framework for choosing products best suited for labeling. L15 © 2012 Elsevier B.V. All rights reserved. M31 Q54

Keywords: Carbon labels Voluntary disclosure Consumer behavior Life-cycle analysis Rebound effect Leakage

1. Introduction retailers to promote green products. Recent examples include Kimberly- Clark's Scott Naturals (household paper products made from recycled Growing consumer interest in “green products” has led many compa- material)2 and Proctor & Gamble's goal of $50 billion in cumulative nies to develop and market products with environmental attributes. In sales of “sustainable innovation products” by 2012.3 2007, the U.S. Patent and Trademark Office saw more than 300,000 appli- A significant and growing proportion of the sustainability-related cations for green-related brand names, logos, and taglines (Ottman, product claims are now focusing on carbon emissions and climate 2011, 35). The number of new products labeled with the word(s) “sus- change. Moreover, carbon labeling of products is gaining considerable tainable”, “sustainability”, “environmentally friendly”,or“eco-friendly” interest. Pilot programs are being implemented in the United King- increased from 100 in 2004 to 526 in 2008, with an additional 450 prod- dom, Switzerland, Japan, and other countries, and there are proposals uct launches in the first four months of 2009.1 Representing a 39% in- to expand and standardize such programs globally (see Vandenbergh crease from the previous year, 6902 products on U.S. shelves in 2010 and Cohen, 2010; Vandenbergh et al., 2011). had some type of environmental claim, including 89 claiming to be carbon neutral (Mintel Group, 2011, Fig. 1). Although the growth of niche brands such as Seventh Generation 2 “ ” and Burt's Bees continues unabated, perhaps even more important Kimberly-Clark Announces Nationwide Launch of Scott Naturals , press release, April 8, 2009, http://investor.kimberly-clark.com/releasedetail.cfm?ReleaseID=375980. is the movement by mainstream consumer product companies and 3 “Proctor & Gamble Deepens Corporate Commitment to Sustainability”, press release, March 26, 2009, http://www.pginvestor.com/phoenix.zhtml?c=104574&p= irol-newsArticle&ID=1270272. The company has since released an even more ag- ⁎ Corresponding author. gressive long-term vision that includes use of 100% renewable energy and 100% re- E-mail addresses: [email protected] (M.A. Cohen), newable or recycled materials for all products and packaging. See “Proctor & [email protected] (M.P. Vandenbergh). Gamble Unveils New Sustainability Vision”, press release, September 27, 2010, 1 “Green Is the New Black”, Adweek, June 24, 2009, http://www.adweek.com/news/ http://www.pginvestor.com/phoenix.zhtml?c=104574&p=irol-newsArticle&ID= advertising-branding/green-new-black-105996. 1475092&highlight=sustainability.

This paper examines the role of carbon product (“carbon foot- ultimately misleads consumers into thinking that the product is print”) labeling in promoting a green economy. While little empirical “greener”, or of higher quality, than it is.5 Darby and Karni (1973) recog- evidence has yet been generated with respect to carbon footprint la- nized this problem and examined the potential role of third-party mon- bels, much can be learned from our experience with similar product itors to evaluate and report on credence qualities. They conclude labels. We first review the theory and evidence on the role of product that if the value of information is high enough to consumers, labeling in affecting product design and consumer behavior. Next, we “high-quality” firms will find it in their interest to disclose this in- consider the role of governments, concluding that whether labeling is formation, and a market will develop for third-party monitors— mandatory or voluntary, either governmental institutions or interna- whether governmental, private, or nonprofit—to verify the truthful- tional, multistakeholder organizations have a critical part to play in ness of this information. As Darby and Karni (1973) note, there is no setting protocols and carbon-labeling standards. We argue that it is inherent reason why this third-party monitor needs to be the gov- important to consider the entire life cycle of a product being labeled ernment. In fact, they conclude that the private (or non-profit) sector and have one international standard for measurement and reporting. could be as effective and potentially more effective than a govern- Finally, we look at the potential impact of carbon product labeling, ment monitor. considering three important issues. First, we examine some of the A related literature examines firm incentives to voluntarily dis- methodological challenges associated with developing credible car- close credence product attributes in product-differentiated markets. bon footprint labels. Second, we consider the potential trade issues As Ippolito and Mathios (1990) show, if one firm can credibly claim associated with carbon footprint labels. Finally, we sketch out an ini- to have a better-quality product (e.g., one that is pesticide free), tial framework for determining which products might be most appro- then the implication to consumers is that competing products with- priate for carbon footprint labeling based on the potential aggregate out that claim do not have that positive attribute. Hence, there is an reduction in carbon emissions. incentive for all firms to compete on that dimension. Ultimately, all firms but the lowest quality will thus have an incentive to disclose. As noted above, the credibility of firms' claims about credence attri- 2. Economic theory of information and product labels butes is a concern. In addition to the direct impact on consumer demand, labeling The economic theory behind the demand for consumer product might bring about an indirect source of demand for lower carbon prod- labels is well established. A traditional utility-maximizing model of ucts through the supply chain. Quality competition among consumer consumer behavior assumes that the rational consumer will choose product firms will provide an incentive for suppliers to innovate and a combination of price and quality that is consistent with her utility offer firms reduced carbon content inputs so that the ultimate product function and constraints. An important assumption of utility maxi- can have a better carbon label. Although we are unaware of any theoret- mization is that consumers have perfect information about the ical literature on this topic, there is ample evidence of downstream price and quality they face. But although consumers easily can de- companies encouraging, assisting and/or pressuring their suppliers to termine quality attributes, or “search” goods, such as color or size, reduce their carbon footprint.6 they do not necessarily observe “credence” goods—the ingredients Baksi and Bose (2007) consider whether self-reporting of credence of a product and their potential harm to the consumer's (or attributes (with governmental monitoring and punishment of cheaters) public's) health—either at the point of purchase or through casual is preferable to third-party certification. Since both are costly, their re- experience (Darby and Karni, 1973; Nelson, 1970).Theroleofprod- sults depend on these costs as well as the relative market share of uctlabelsisthustoturna“credence” attribute into a “search” attri- “green” versus “brown” firms. They conclude that self-reporting is pref- bute so that consumers easily can compare and make more erable when the market share of “green” firms is large enough that any informed (utility-maximizing) decisions. loss from cheating by “brown” firms is outweighed by the cost of mon- For consumers, if the value of additional information exceeds the itoring by third parties. A more realistic scenario, however, is that the cost of search, they will demand this information and use it in their pur- potential cheating by “brown” firms will make mandatory third-party chase decisions. Consumer demand for information on credence attri- certification for all firms a more preferable outcome. butes is also predicated on the assumption that consumers know this attribute exists and might vary by product. In other words, in order to 2.1. Products with public and private attributes demand information, consumers need to know the value of it. The supply side of the market is more complex. What are the incen- While the early information disclosure literature envisioned attri- tives for firms to supply this information? Firms whose products pos- butes that directly benefit consumers, such as nutrition labeling or sess a credence attribute with a positive benefit to consumers have an health risks, consumers also may demand credence goods that have incentive to disclose and thus turn this latent value into a search attri- a “public” good component, such as environmental protection. Con- bute that presumably will increase sales. Profit-maximizing firms will sumer demand for such attributes may arise either from altruism or thus weigh potential increased revenues against the cost of testing the “warm glow” associated with spending money on environmental and providing information. In addition to directly increasing sales protection (Andreoni, 1990). through higher consumer demand, however, firms also might benefit Firms also have an incentive to disclose these public attributes of from improved relationships with regulators and other stakeholders their products in some cases. Following the earlier work on voluntary that might positively impact profits in other ways. disclosure, Ibanez and Grolleau (2008) consider the case of eco-labels If there is no cost to disclosure and consumers can verify/trust and show that as long as it is more costly for the “polluting” firm to ob- these disclosures, all sellers will disclose their quality for fear that tain the label than the “clean” firm, the latter will obtain the label, and consumers will otherwise infer that no disclosure means “lowest quality” (Grossman, 1981).4 However, if it is costly to verify claims, fi rms have an incentive to claim that their products are high quality. 5 Lyon and Maxwell (2011, 9) define greenwashing as “…fundamentally about mis- This might take the form of fraudulent claims or more subtle attempts at leading consumers and investors by telling the truth, but not the whole truth. This sug- “greenwashing”, whereby firms selectively disclose information that gests a model in which the firm discloses verifiable information, but may choose to withhold facts that do not reflect favorably on it, thereby persuading outsiders that the firm's performance is better than it is in reality.” 6 Many large global consumer product companies and retailers impose supply chain 4 See Dranove and Jin (2010) for an extensive review of the literature on quality dis- requirements such as environmental (including carbon) reporting and/or manage- closure and certification. ment systems. See for example, Vandenbergh (2007). M.A. Cohen, M.P. Vandenbergh / Energy Economics 34 (2012) S53–S63 S55 pollution will be reduced. However, as Ibanez and Grolleau caution, un- collective action and help to overcome the free-rider problem in- less consumers are all altruists, eco-labels will be only a partial solution herent in individual purchase decisions involving a public good. to environmental externalities. For example, pressure by an environmental group and the threat Many environmental attributes will have private and public value. of boycotts might induce firms to improve the environmental foot- For example, products that use less toxic chemicals might have both print of their products—or of products they put on their shelves— direct health benefits as well as benefits to the environment. One im- even in the absence of direct consumer demand signals (Baron portant aspect of products that have both types of value is that the and Diermeier, 2007). Lenox and Eesley (2009) provide evidence consumer oftentimes has a choice of purchasing this bundled product on the role of nongovernmental organization boycotts and other or purchasing the products separately. For example, instead of purchas- forms of pressure on firm behavior. ing a “green” product, a consumer might purchase a less expensive Thus, the demand for labeling of products may not necessarily product without an environmental benefit and instead donate to an en- come directly from consumers. Indeed, in some cases, firms might vironmental cause. Kotchen (2005, 2006) examines this case and pro- lead as opposed to follow consumer demand on issues such as carbon vides some interesting insights into the way in which consumer footprints. For example, firms might be concerned about the long-term product choice and environmental protection interact. It is possible, cost or viability of certain inputs due to climate change and future reg- for example, that the availability of new “green” products will reduce ulatory responses to climate change, and thus might pressure suppliers donations (or other expenditures on environmental protection) as con- to reduce their carbon footprint (see e.g., Lyon and Maxwell, 2011; sumers shift their discretionary budget away from donations and into Roberts, 2011; Vandenbergh, 2007). As examples, Wal-Mart announced consumer products with environmental attributes that are more expen- a plan in 2010 to assist its suppliers in reducing their carbon footprint by sive. This has been termed the “rebound effect”. 20 million metric tons by 2015—citing the expectation of higher energy On the other hand, it is possible that the availability of new prices in the future as a driving factor,7 and P&G announced a goal of “green” products will provide information to consumers about the 100% renewable energy use in its manufacturing processes based on relative importance of certain environmentally friendly behaviors their responsibility as a world's largest consumer package goods com- and thus increase the demand for environmental protection expendi- pany.8 Neither of these companies talked about consumer demand as tures through either donations or political pressure (Kals et al., 1999; a driving force behind these commitments. Maiteny, 2002). In other words, it is important to know whether Third-party eco-labeling and certification organizations—often- “green” products are substitutes for or complements to other envi- times with competing goals and conflicting interests—have proliferat- ronmental protection activities. ed in recent years. Ottman (2011, 165) reports there are currently more than 400 different eco-labels or certification schemes in 207 2.2. Bandwagon effects and the “status” of being carbon neutral countries. Third-party verification is only as good as the third-party itself. Some of these groups are industry sponsored, while others are in- Economists recognized many years ago that consumer demand for dependent and subject to rigorous and transparent multistakeholder products might be driven by “status” associated with consumption. scrutiny. This proliferation easily can lead to consumer confusion and As Liebenstein (1950) explained, while status demand is often associ- frustration because it requires expertise to know the difference be- ated with conspicuous consumption (i.e., products are demanded between paper labeled with the “Forest Stewardship Council” versus the cause of their high price and the signal it sends about one's social “Sustainable Forest Initiative”, for example. The Natural Resources De- status), it may also be associated with either the “bandwagon” or fense Council has attempted to collect these various labels (and less rig- “snob” effect. The bandwagon effect is one of conformance—based orous marketing claims) and provides its own ranking of third-party on the desire to be associated with a certain social group or to be fash- labels.9 Organizations also have formed to set industry standards for ionable or stylish, while the snob effect is just the opposite—the de- voluntary product certification schemes.10 sire to be different from the masses. Having a low carbon footprint might be desirable for either reason depending upon the nature of so- 3. Evidence on green labels cial norms in a society at a particular time. In fact, it is possible that what starts out as ‘snob’ appeal by a few early adopters or trend setters Studies on the effectiveness of green labels generally do not sat- can at some point cascades into a ‘bandwagon’ appeal and becomes the isfy the standards of rigorous empirical research because they lack social norm (see e.g. Heal and Kunreuther, 2010). While the evidence is random assignment or quasi-experimental designs. As a result, largely anecdotal (see e.g., Vandenbergh and Steinemann, 2007), the they fail to establish the causal relationship between green labels fact that major league sports teams, mayors, university presidents, and environmental outcomes. Instead, the nature of the evidence and even rock groups are moving towards carbon neutrality suggests falls into two categories: industry and market studies of product that this is a possibility in the case of carbon footprints. sales; and consumer surveys of label awareness, use, and stated preferences. 2.3. Role of third parties 3.1. Industry and market studies As noted above, third-party monitors are critical in information markets because they allow firms to credibly disclose product infor- Perhaps one of the most studied programs is the U.S. Energy Star label, mation that consumers might not otherwise believe. However, third apublic–private partnership in which the U.S. Department of Energy cer- parties may serve other important purposes in information markets. tifies consumer products that meet certain energy-efficiency criteria.11 Organizations such as Consumer Reports develop the expertise to an- According to the 2009 Residential Energy Use Survey conducted by the alyze and report on product attributes using their own judgment and U.S. Energy Information Administration, market penetration of Energy criteria. Retailers also may adopt a similar role in screening which products to carry or which to label the most environmentally friend- 7 See Stephanie Rosenbloom, “Wal-Mart Unveils Plan to Make Supply Chain ly. Instead of doing considerable research to understand and com- Greener”, New York Times, February 25, 2010, B3. pare product attributes, consumers may rely on these sources to 8 See http://www.pg.com/en_US/sustainability/environmental_sustainability/index. rank products. shtml. 9 See http://www.nrdc.org/living/labels/. In the context of a public attribute or externality, third parties 10 One example is ISEAL; see www.isealalliance.org. might play a different role of aggregating consumer interests. These 11 The Energy Star label generally applies to products with energy usage that is at third parties might reduce the transactions costs associated with least 25–30% below mandatory requirements. S56 M.A. Cohen, M.P. Vandenbergh / Energy Economics 34 (2012) S53–S63

Star appliances is about 40% for refrigerators, dishwashers, and clothes future labeling program. It also highlights the fact that consumer pur- washers.12 The U.S. Environmental Protection Agency claims that in chase decisions might impact the marginal consumer and that analyz- 2006 the Energy Star program helped to reduce 37.6 million metric ing market segments is important in understanding the ultimate tons of carbon emissions, although an Inspector General report has impact on consumption patterns. questioned the accounting methods used to generate the agency's esti- An experiment in a grocery store in Australia labeled six product cat- mates (U.S. EPA, 2008). Brown et al. (2002, 515) estimates that the Ener- egories as having either a “below average”, “near average”,or“above av- gy Star label in the U.S. has reduced carbon emissions by 150 million erage” carbon footprint relative to competing brands (Vanclay et al., metric tons between 2001 and 2010, an amount that “represents about 2011). Overall, the categories that were below average lost 6% of sales, 4% of carbon emissions for the residential and commercial sectors over while those that were above average gained 4%. However, there was con- thesameperiod”. siderable variation depending on price differences. When the green- All these estimated reductions are based on the market penetra- labeled products were also cheapest, they gained 20% of sales following tion of Energy Star labels and engineering-based estimates of product the introduction of the labeling program. usage and emissions. Thus, while we know that Energy Star appliances In the United States, there is evidence that a significant percentage have lower energy usage and lower carbon emissions than competing of customers are willing to pay a higher price for “green” renewable products, we do not know whether these products would have been electricity when given that option on their electricity bills. In 2009, ap- manufactured and purchased in the absence of the label. While it is like- proximately 1.4 million U.S. customers purchased green energy (Bird ly that Energy Star has spurred manufacturers to innovate and improve and Sumner, 2010, 7–8). Of these, residential customers purchased the energy efficiency of their products, we cannot necessarily attribute 7.2 million megawatts of green electricity, while the nonresidential sec- 100% of any estimated energy-efficiency benefits to the program itself. tor purchased 22.8 million megawatts. While impressive, these figures Thus, these estimates are best thought of as an upper bound. currently represent less than 1% of total electricity consumption in the The program also might be a best-case scenario in terms of con- United States. However, participation varies widely by program. The av- sumer adoption. Because Energy Star products lower carbon emis- erage utility green pricing program reportedly has about a 2% participa- sions by reducing energy consumption, they provide a measurable tion rate, but top-performing programs achieved rates ranging from consumer benefit in the form of less expensive energy bills. While 5.1% to 20.8% (Bird and Sumner, 2010, 9). the Energy Star label offers this significant private benefit in addition Kotchen and Moore (2007) studied participation in two green to any potential warm glow, altruism, or status benefit, many other electricity programs and found that participation is higher with in- consumer products, such as food and cosmetics, will not. For example, come level, environmental concerns, and altruistic attitudes. Their a comparison of frozen versus fresh orange juice finds that frozen findings suggest consumer segments and marketing approaches juice has a lower carbon footprint due to its lower volume, requiring that might increase the uptake of green electricity programs (for ex- less transportation and refrigeration space.13 To a consumer, price ample, a joint marketing campaign with environmental and/or char- and quality of orange juice likely will be of most interest—and any im- itable organizations). pact of a carbon footprint label on the type of orange juice consumed most likely will have to come through warm glow, altruism, or other 3.2. Consumer surveys external pressure on orange juice manufacturers or distributors. Aside from Energy Star, numerous anecdotal industry case studies In the previous section, we discussed industry and market studies demonstrate the potential for consumer labels to have a significant that examine actual purchase behaviors to uncover consumers' will- effect on products sold in the market. Perhaps the most widely ingness to pay for environmental amenities. Researchers have also known example is dolphin-safe tuna. Public concern about killing dol- used surveys to elicit preferences through careful random design. phins in the process of catching tuna led to a significant drop in tuna For example, a study by Borchers et al. (2007) provided respondents demand. In this case, “dolphin-safe tuna” labels helped to revive the with several hypothetical bundles of increased electricity costs and canned tuna market (Teisl and Hicks, 2002). Interestingly, part of green energy sources (solar, wind, biomass, farm methane, and a ge- the shift toward the sale of dolphin-safe tuna apparently came as a re- neric green source), and asked them to choose between these differ- sult of direct pressure by environmental groups on retailers. Thus, ent options or continue current energy sources at no additional cost. third-party intermediaries may have played as much (if not more) They found a positive willingness to pay for green energy, with the of a role in changing this market as did direct consumer purchase de- highest value being placed on solar energy. cisions.14 This is consistent with other evidence on the role of Not all consumer surveys are based on hypothetical purchases. third-party nongovernmental pressure on firm environmental behav- Clark et al. (2003) surveyed consumers who purchased green power ior (see, e.g., Lenox and Eesley, 2009). (as well as those who did not) in the Detroit, Michigan area. They A panel study of actual purchase behavior for about 1600 con- found that altruism toward the environment was the most significant sumers between 1997 and 2001 in Denmark found that purchasers factor in choosing to purchase green power, followed by altruism to- of toilet paper with a Nordic Swan eco-label had an increased margin- ward regional residents and individual health-based concerns. While al willingness to pay of 13–18% (Bjørner et al., 2004). Interestingly, there was some evidence of warm glow, it was the least important Bjørner et al. found little evidence of a higher willingness to pay for reason. They also found that local health concerns outweighed global paper towels with the Nordic Swan label. They conjecture this result concerns. arises because cloth is an even more environmentally friendly substi- Although studies have found evidence of altruistic behavior on tute for regular paper towels than the certified ones are, and thus the the part of consumers, there is limited and often conflicting evidence most environmentally conscious consumers are less likely to buy any of consumer willingness to pay for carbon emission reductions be- paper towels. This example highlights the importance of considering yond pure energy savings or other personal benefits. For example, available product substitutions when considering the impact of any Jacobsen et al. (2012) report on a green power program in Memphis, Tennessee, where consumers were given the option of paying $4 per 12 Data compiled from 2009 Residential Energy Consumption Survey, Table HC3.2, month for green energy blocks. At the end of 2008, the reported up- available at http://www.eia.gov/consumption/residential/data/2009/. Of those house- take was less than ½ of 1% of customers—although other green holds who report that they have a refrigerator, 37.3% claim it is Energy Star. The corre- power programs in the U.S. have reported levels as high as 3% or sponding figures are 40.3% for dishwashers and 44.0% for clothes washers. more (Wiser et al., 2000: 12). In contrast, Michaud et al. (forthcoming) 13 See http://www.carbon-label.com/our-news/case-studies/tesco. 14 See Vandenbergh and Cohen (2010, 276) for a discussion of this phenomenon and conducted a random discrete choice experiment in Grenoble, France the evidence in the case of tuna. where consumers faced actual purchase decisions of roses. Irrespective M.A. Cohen, M.P. Vandenbergh / Energy Economics 34 (2012) S53–S63 S57 of price, the “low carbon” rose was chosen 79% of the time compared to greenhouse gas emission reductions in their home country will shift a “higher carbon” rose. Moreover, consumers were found to be willing production to other countries with less stringent requirements, thereby to pay a significant price premium for the low carbon rose. Since a reducing the impact of any climate legislation. In theory, it is possible low carbon rose has no direct consumer health benefits, it would be rea- that leakage could result in even worse emissions in some cases because sonable to attribute this demand to altruism (or perhaps to the “status” the production technologies employed could be dirtier than existing of bringing carbon friendly roses to a loved one). As both economic the- pre-regulated methods in the home country, and longer transportation ory and these two examples suggest, consumer demand for low carbon distances increase the carbon footprint of bringing these newly imported products is likely to vary by product category, consumer income, tastes, goods into the home country (Wiedmann et al., 2008). In fact, Watson and consumer preferences for climate risk reduction—all of which and Moll (2008) have estimated that leakage and offshoring of produc- might vary considerably by location. tion have accounted for virtually all the United Kingdom's reduced car- Although not a market test of actual consumer behavior, a recent bon emissions from 1990 to 2005, and that if carbon emissions were survey found that 30% of consumers indicated they would be willing calculated throughout the supply chain, instead of a 15% reduction, the to pay more for a product from a good corporate citizen (Mintel country's emissions have actually increased by 19%. None of this speaks Group, 2011, Fig. 25). In contrast, 53% of respondents claimed that to the more potent political concerns of losing jobs in the home country. “good corporate behavior” would affect their own purchase behavior Thus, leakage is a very real concern that must be addressed in any serious if “price, quality and convenience are equivalent to competing prod- proposal for greenhouse gas regulation. In the context of greenhouse gas ucts”. Since specific, verifiable product claims such as the carbon con- regulations, many of the proposals to deal with leakage involved taxing tent of a product are more likely to elicit positive consumer behavior or subsidizing goods that are traded internationally.16 than general claims about corporate citizenship, these results might In theory, a carbon label may prompt similar leakage concerns. For be an underestimate of the underlying demand for carbon labels. example, if a country requires carbon labels on products produced lo- Moreover, as companies increasingly find more cost-effective ways cally, there might be an incentive to offshore production and import to reduce life-cycle carbon emissions without raising product prices, more products without a carbon label. Of course, as we mention in trade-offs between product prices and carbon emissions are likely Section 2, economic theory suggests that consumers will infer that to be less important. Indeed, as one corporate executive noted, “Con- products that are not labeled are necessarily higher in carbon content— sumers expect products that give brilliant results, at an affordable thus mitigating against this concern somewhat. In practice, however, price, and deliver sustainability benefits—i.e., no trade-offs” (White, we would not anticipate countries adopting carbon labels only for do- 2009, 389). mestically produced goods. Instead, in the absence of a global standard, Researchers also have used surveys to examine consumer aware- we would expect countries to adopt a labeling requirement on all prod- ness of environmental labels, finding that awareness varies dramati- ucts sold within its borders (Vandenbergh and Cohen, 2010). cally by type of label. For example, a 2009 U.S. survey reports that Thus, carbon footprint labels may help alleviate some of the leakage awareness of the Energy Star label is as high as 93%, with 73% of con- concerns that arise with other carbon emission regulatory schemes. sumers claiming they are more likely to purchase a product with the First, let's consider the direct effect of a carbon footprint label. Presum- label (Ottman, 2011, Fig. 7.2). In fact, 34% of respondents claimed to ably, the existence of the label will induce producers to lower their car- have purchased energy-efficient electronics or appliances within the bon footprint so as to attract consumers who demand this product past three years (Ottman, 2011, Fig. 1.3). In contrast, only 18% are attribute. If the carbon footprint label is a global standard and is appro- aware of the Marine Stewardship Council label for sustainable fish, priately enforced and complied with, one international trade incentive and only 10% claim they would likely base a purchase decision on created by carbon labeling would be to move production from one the label. country to another based on the availability of lower carbon fuel sources Awareness of eco-labels varies significantly by country; yet expe- or production methods. At the very least, this will reduce the incentive rience has shown that consumer awareness can grow dramatically in to move production overseas to “dirtier” facilities. It also might provide a very short period of time. For example, the Rainforest Alliance an incentive for producers in the unregulated country to reduce their reported that consumer awareness of their green frog seal grew greenhouse gas emissions voluntarily in response to increased consum- from 26% to 35% between 2009 and 2010 in Germany, and is as high er demand pressures from abroad. Similar effects are likely in the ab- as 54% in the United Kingdom and Ireland.15 sence of a global standard as long as the requirements apply to all products sold within the regulated country. While the direct effects of carbon labeling are clearly in the direction 4. Potential impact of carbon labels of reducing carbon emissions, consideration of potential feedback effects complicates the picture. For example, Swallow and Sedjo (2000) consid- In this section, we consider the potential impact of carbon labels er eco-labeling of forest products in a general equilibrium context and on aggregate greenhouse gas emissions. As discussed above, there is show that reduced demand for forestry products possibly will induce ample evidence that consumer purchase decisions (whether directly the conversion of marginal forestry land to less ecologically desirable ag- or indirectly through retailer actions) will respond at the margin to ricultural production. Thus, once general equilibrium and cross-product credible claims that certain products have environmental benefits substitution effects are considered, the implications of carbon labeling over others. However, the question still arises whether carbon label- are not entirely clear. ing can be expected to bring about any meaningful reduction in emis- To take another example, it is possible that labeling the carbon foot- sions. We address three of the most widely articulated concerns: print of beef in a developed country could reduce the demand for beef leakage outside any labeling scheme, the “rebound” effect, and the and lower its price sufficiently to induce an increase in the demand potential magnitude of carbon reductions. for beef in a developing country. The net result might be less carbon reduction than anticipated—or in the worst case, an increase in carbon 4.1. Leakage emissions if beef is now transported internationally more than previ- ously. Thus, careful empirical analysis of the cross-price elasticity of The concern over leakage has been one of the most widely discussed products and differential price elasticities across countries (or even con- issues when debating cap-and-trade legislation in the United States and sumer segments within a country), industry structure, and alternative elsewhere. In particular, policymakers are concerned that requiring 16 See, e.g., Stavins, 2008 for a discussion of leakage and Fischer and Fox, 2011 for an 15 See www.rainforest-alliance.org/marketing/why. analysis of alternative proposals to reduce leakage from greenhouse gas regulation. S58 M.A. Cohen, M.P. Vandenbergh / Energy Economics 34 (2012) S53–S63 uses of production inputs is warranted before definitive policy recom- Danish consumers over a three-year period found that heavy recy- mendations can be made. clers tended to increase their purchase of organic foods, although Although unilateral action by one country might bring about re- the researchers observed some potentially offsetting behavior ductions in the carbon footprint of its home country consumers, the (Thøgersen and Ölander, 2003). This remains an important topic absence of a global standard brings with it new complications. In par- for future research. ticular, unilateral action of this nature raises the potential of trade wars. It also highlights the importance of an objective and standard 4.3. Can consumer product labeling really have an impact? methodology that is not designed to aid one country's products over another. These issues are addressed in Section 5. The production, transport, and consumption of consumer goods account for a substantial share of U.S. and global greenhouse gas emissions, 4.2. Rebound effect versus increasing social norm for carbon reductions suggesting that small changes in consumption may have important effects on domestic and global emissions. In 2005, direct energy use by In addition to “leakage” across countries, concern has been expressed U.S. households accounted for approximately 38% of overall U.S. carbon that individual consumers might offset their own green purchasing deci- dioxide (CO2) emissions, or 626 million metric tons of carbon (Energy sions by reducing their environmentally friendly behavior in another di- Information Administration, 2008; Gardner and Stern, 2008). The total mension of their lives. One can construct a rational utility-maximizing from household energy use represents approximately 8% of global emis- model of consumer behavior whereby lowering the search costs through sions, an amount that is larger than the total emissions of any country carbon footprint labels provides new information to consumers that al- other than China (Dietz et al., 2009; Vandenbergh and Steinemann, lows them to rebalance their mix of low-carbon and high-carbon 2007). Consumer goods potentially subject to carbon labeling account products. For example, it is possible that a consumer who currently for much of the energy use and emissions from the U.S. household sector purchases “green power” through their electricity provider will now (Dietzetal.,2009,Table1). feel better about her use of electricity and will increase usage, par- To the extent carbon labeling reduces leakage arising from inter- tially (or even fully) offsetting the emission reductions from pur- national trade, it has the potential to affect a large share of global chasing green power. Alternatively, a green power purchaser who greenhouse gas emissions. One study concluded that leakage from learns about a newly instituted carbon-labeling scheme might re- the United States exceeded 20% of global warming potential in 2004 duce her green power consumption and use that savings to pur- (Ghertner and Fripp, 2007). A consumption-based model concluded chase a low-carbon consumer product that costs a little more than that goods and services traded internationally accounted for 23% of the consumer product she used to buy. This consumer will ultimate- global CO2 emissions in 2004 (Davis and Caldeira, 2010), and the ly be better off from carbon labeling, but her carbon footprint might share attributable to internationally traded goods and services appears remain unchanged. These are examples of what has been termed the to be increasing. For example, Peters et al. (2011) concluded that emis- “rebound effect”. sions from the production of internationally traded goods increased

There is considerable evidence on the rebound effect as it re- from 20% of global CO2 emissions in 1990 to 26% in 2008. lates to energy efficiency enhancements. For example, a review by The growing recognition of the importance of emissions embodied Greening et al. (2000) found a rebound effect of between 0% and in trade has generated calls for shifting national emissions inventory ac- 50% over 75 studies in the U.S. on everything from automobiles, counting from an exclusive focus on emissions within national bound- residential lighting, water heaters and space heaters, while Sorrell aries toward a consumption-based approach (Peters and Hertwich, et al. (2009) finds that no more than 30% of any home energy effi- 2008). ciency improvements are likely to be offset by increased energy usage in OECD countries. 5. Practical issues There is little evidence to date on any rebound effect from voluntary purchases of carbon emission reductions. However, the evi- The goal of carbon footprint labels is to provide businesses and con- dencetodatesuggeststhatwhilethereissomeoffsettingbehavior, sumers with meaningful information that will ultimately allow them to it is unlikely to offset carbon reductions in any meaningful way. For make informed decisions about product choice, use, or both. To achieve example, Jacobsen et al. (2012) studied electricity consumption for that goal, numerous practical issues must be overcome. This section consumers in Memphis, Tennessee, some of whom had purchased briefly reviews some of the major methodological concerns, interna- green power at a premium price. In the aggregate, they found no sta- tional trade issues, and design issues associated with carbon labeling. tistically significant difference in energy consumption following pur- Critics, including some environmentalists, claim that these challenges chase of green energy. However, households at the minimum level of make carbon labeling misleading and potentially unable to change con- green energy participation did increase their electricity consumption sumer behavior.17 While these concerns are potentially valid, the key by approximately 2.5% after enrolling in the program. Even for these question is whether carbon footprint methodologies and standards consumers, however, the net effect on carbon emissions was actually can be developed to minimize these issues—and at what cost? negative because any increase in emissions due to their increased consumption was more than outweighed by the reduced emissions 5.1. Methodological challenges from their green energy purchase. While the Jacobsen et al. results suggest only minimal offsetting The methodological challenges to implementing a reliable carbon- behavior—and a significant positive impact on energy consumption labeling program are significant. Measuring and verifying the carbon (and a net decrease in emissions)—we do not know if the green emissions of a product's life cycle involves numerous assumptions and power consumers increased or decreased their consumption of other compromises. For example, a product might be manufactured at several products with a carbon footprint. On the one hand, it is possible that different facilities, each using different energy sources and having differ- some of these consumers increased their consumption of other “dirty” ent paths of shipment to their final destination. Thus, two identical prod- products upon deciding to purchase green power. On the other hand, ucts might have different manufacturing carbon footprints. At the other it is quite possible that the effect would go the other way—heightened end of the life cycle, consumers differ in how they use and dispose of awareness and habits of purchasing carbon-friendly products might in- the product. crease consumer propensity to reduce their carbon footprint elsewhere in their life choices. Although there is little or no evidence on this 17 For example, see Julia Hailes, “Carbon Footprint and Carbon Labeling,” http://www. effect in the context of greenhouse gas emissions, one study of juliahailes.com/pdfs/CarbonFootprint-PrintedVersion.pdf. M.A. Cohen, M.P. Vandenbergh / Energy Economics 34 (2012) S53–S63 S59

Koning et al. (2009) provide an example of the uncertainties associat- trade treaties may serve as barriers to a carbon labeling system: the Gen- ed with comparing the carbon footprint of ultra-liquid versus compact eral Agreement on Tariffs and Trade (GATT) and the Agreement on Tech- powder laundry detergent. In their example, they show how increasing nical Barriers to Trade (TBT Agreement). A carbon-labeling program may the discretion allowed for the choice of life-cycle model parameters can take the form of a mandatory public system, voluntary public system, or result in misleading results. For example, when key assumptions, such voluntary private system, and the vulnerability to a trade challenge as end-user washing-machine efficiency, temperature selection, and under these treaties likely declines in that order. electricity sources, are standardized, ultra-liquid detergent results in a In general, the GATT prohibits discrimination against the import of lower carbon footprint 100% of the time. When most of these parameters “like products”. For example, import laws cannot discriminate in favor are allowed to vary, however, the compact powder detergent is estimat- of domestic goods over “like products” from other nations. Jurisprudence ed to have a lower carbon footprint 23% of the time. developed in some trade disputes typically defines “like products” in The methodological issues associated with carbon footprint label- terms of the physical characteristics of goods as opposed to the process ing give rise to numerous policy challenges. On a practical level, they and production methods (“PPMs”) used in the harvesting, processing, highlight the importance of adopting a single, globally recognized manufacture or production of the goods (Kysar, 2004; Vranes, 2011). protocol for standardized carbon footprint methodologies. Two or- This focus on the physical characteristics of the good rather than PPMs ganizations have developed such life-cycle protocols using similar may induce the trade regime to view a good with very different prove- principles and a multistakeholder approach. These protocols are nance from another to be a “like product” (e.g., GATT Arts. I(1) and the British Standards Institution (2011) Publicly Available Specifica- III(4)). A life-cycle analysis-based government carbon-labeling system tion 2050 and the Greenhouse Gas Protocol's (2011) Product Life thus may be vulnerable to a claim that it is a PPM measure that discrim- Cycle Accounting and Reporting Standard. The International Stan- inates against “like products”, even though the products have different dardization Organization also is developing a carbon-labeling stan- carbon footprints. Under Article III of the GATT, regulatory measures dard, ISO 14067, expected to be finalized within the next year or that treat like products differently based on non-product-related PPM is- two.18 sues (often referred to as “npr-PPMs” or “unincorporated PPMs”)maybe According to the Greenhouse Gas Protocol, the two existing viewed as discriminatory even if the measures are neutral as to country standards are very similar and unlikely to result in significant dif- of origin (Joshi, 2004). ferences in their measurement outcomes. Both provide principles The npr-PPM measures, if deemed discriminatory and a violation of on how to develop transparent life-cycle measures for products. the GATT, may nonetheless meet one of the exceptions provided in However, they are not detailed enough to provide aggregation GATT Article XX (Charnovitz, 2002; Pauwelyn, 2004; Vranes, 2011). rules or sector-specific assumptions. Instead, further work at the Provided that the measures “are not applied in a manner which would sector (and even product category) level is needed to arrive at con- constitute a means of arbitrary or unjustifiable discrimination … or a sistent, comparable carbon footprint labels. disguised restriction on international trade”, Article XX permits Even if a standardized methodology is adopted, however, discrep- measures that are otherwise illegal if they are “necessary to protect ancies inevitably will arise between the carbon footprint label and the human, animal or plant life or health” or, a somewhat less stringent emissions realized by one consumer (or class of consumers). For ex- test, “relating to the conservation of exhaustible natural resources ample, the extent to which an all-electric vehicle produces fewer if such measures are made effective in conjunction with [similar do- greenhouse gases than a high-efficiency gasoline-powered vehicle mestic measures]”.20 But there is very little case law on labeling or depends greatly on the electricity fuel mix, which can include low- other climate-related questions, and whether WTO dispute bodies or high-efficiency coal-fired power plants, natural gas‐powered elec- would construe Article XX to permit carbon-labeling systems is tricity, renewable energy sources, and the like.19 Depending on the unclear (Brazil—Tyres; United States—Shrimp/Turtle I). frequency and magnitude of this discrepancy, one can envision vari- The possibility that a discriminatory trade restriction will be deemed ous possible solutions: exempt, and thus survive a trade challenge, diminishes in situations where the environmental effects concern another country's environ- • do nothing (i.e., maintain one carbon footprint label); mental quality or resources, or where the basis of the restricting • prepare different labels depending on the local source of electricity; country's concern about global environmental resources is viewed • develop a more complex label that provides multiple values that as illegitimate. For example, Mexico challenged U.S. dolphin-safe la- depend on the local source of electricity; beling restrictions in part by asserting that they were based on ani- • or determine that the high degree of variability and lack of clear su- mal rights concerns for non-endangered dolphin populations and periority among products are such that carbon labels for this prod- thus more trade restrictive than necessary to achieve a legitimate ob- uct category should not be a priority. jective (United States—Tuna/Dolphin III Panel Report).21 On the other hand, if there is an adequate nexus between resources in 5.2. Trade issues each of the countries involved (e.g., turtles that move between the waters of the nation adopting the trade restriction and the restricted na- The international trade regime poses a potential barrier to some tion (United States—Shrimp/Turtle I and II)), or if the environmental types of product carbon-labeling systems. The World Trade Organization concern is a matter covered by an international treaty (e.g., sea turtles (WTO)overseesacomplexsetofrulesthataredesignedtostrikeabal- in United States—Shrimp/Turtle I and the Convention on International ance between the right of WTO members to advance legitimate goals, in- Trade in Endangered Species), the trade restriction may be upheld. cluding environmental protection, and the right of other members not to have such measures applied arbitrarily or serve as a form of disguised protectionism (Dunoff, 1994). Two of the organization's international 20 The Shrimp/Turtle II ruling makes it clear that process-based measures may fall within the scope of the general exceptions clause (United States—Shrimp/Turtle II). 18 The Greenhouse Gas Protocol is a cooperative initiative of the World Resources In- 21 The WTO's most recent treatment of labels, the May 2012 report of the Appellate stitute and the World Business Council for Sustainable Development. The ISO 14067 Body in Mexico's appeal of the Tuna–Dolphin III panel decision (United States—Tuna/ standards are being developed in collaboration with both organizations. See “Quantify- Dolphin III AB Report), was issued as this article was going to press and is not reviewed ing the Greenhouse Gas Emissions of Products: PAS 2050 & the GHG Protocol Product here. Although the May 2012 Appellate Board decision reversed the Tuna–Dolphin III Standard,” http://www.ghgprotocol.org/standards/product-standard. panel decision, it upheld the panel's determination that the purpose of the U.S. labeling 19 This point has been made in the case of China's recent push to adopt electric vehi- program was legitimate, and the reversal was based on reasoning that does not affect cles. In many parts of the country, there might be little or no carbon benefit from an the conclusion that private, voluntary labeling systems are less vulnerable to challenge electric vehicle under current electricity generation technologies (Earley et al., 2011). than public, mandatory labeling systems. S60 M.A. Cohen, M.P. Vandenbergh / Energy Economics 34 (2012) S53–S63

It is not clear how these issues will be addressed for an npr-PPM suggest that the definition of “mandatory” may be viewed broadly, and that has a global or universal impact, such as a carbon-labeling sys- the status of even voluntary government-administered npr-PPM mea- tem. Carbon emissions externalize harm in ways that may affect the sures under the TBT Agreement is hotly debated (Joshi, 2004; Vranes, restricted country and the country adopting the carbon-labeling mea- 2011). sure, and they may affect not just abstract environmental concerns In addition, the approach of the trade regime to private carbon- but physical environmental conditions. Carbon emissions in China labeling systems may differ from the approach to government sys- arising from a good produced for a Danish consumer are likely to af- tems. A robust literature and several WTO dispute reports exist on fect Denmark as well as China, and the potential harms arising from mandatory and voluntary government environmental measures, changes in climate and sea level in both countries are more concrete but many eco-labeling programs are developed and administered than concerns about global dolphin populations. The existence of the by nongovernmental organizations and private firms. The trade Framework Convention on Climate Change and its Kyoto Protocol, implications of these private programs, which may have little or whose purpose is to reduce and restrain carbon emissions to avoid no governmental involvement, have been studied less exhaustive- such potential harms, may buttress this point with policymakers. ly, but there are reasons to believe that they are less vulnerable to A WTO dispute settlement panel may evaluate a TBT Agreement aWTOchallenge. claim before turning to a GATT claim (EC—Asbestos; United States— Although only states can be parties to WTO disputes, a private Tuna/Dolphin III, Panel Report).22 The TBT Agreement extends to all organization's labeling program can give rise to a WTO dispute. technical regulations, standards, and conformity assessment procedures The TBT Agreement obliges WTO members to take “all reasonable that apply to trade in goods. Under the TBT Agreement, a labeling system measures” to ensure that private standardizing bodies adhere to a determined to be a “technical regulation” must be neither discriminatory Code of Good Practice found in an annex to the TBT Agreement. nor “more trade-restrictive than necessary” to fulfill a legitimate objec- Moreover, acts by private parties can give rise to disputes when tive and should be based on an international standard, if one exists these acts are “attributable to a WTO member”.26 The boundaries (TBT Agreement 2.1, 2.2).23 Prior panel decisions suggest that the WTO of what is attributable to a WTO member are unclear, but one takes a broad view of both “technical regulations” and “more trade- WTO panel determined that private actions with “sufficient govern- restrictive than necessary” (EC—Sardines; United States—Tuna/Dolphin ment involvement” may be attributed to a WTO Member (Japan— III Panel Report). Although technical regulations that are related to the Film, paragraph 10.52). The panel declined to elaborate on the defini- product clearly fall within the scope of the TBT Agreement, substantial tion of sufficient government involvement,27 but some combination of disagreement exists over whether technical regulations are subject to actions, such as government funding of the standard development or the agreement if they are not related to the product's characteristics implementation, government approval of the standard or the partici- (Joshi, 2004; Vranes, 2011). In addition, in the recent Tuna/Dolphin dis- pants in the standard setting, or inclusion of the standard in procure- pute, Mexico claimed that government labeling standards were tanta- ment,28 could raise concerns (e.g., Argentina—Leather; Canada—Dairy; mount to a ban on the sale of non-dolphin-safe tuna in the United EC—Apples; Japan—Film). In short, although mandatory and voluntary States, although a label that merely provides information (e.g., a carbon government labeling programs may avoid or survive a WTO challenge, footprint score) may not be viewed in the same light as the dolphin- a private approach with limited government involvement may be the safe label standards. A number of current government-sponsored eco- least vulnerable of the three (Vandenbergh et al., 2011). label programs provide PPM information, and the WTO's Committee on The ultimate effects of the trade regime on carbon labeling also may Trade and Environment has been studying eco-labels for over a decade be influenced by how these issues are framed in the broader policy con- without taking a stand on these labels. text. Carbon labeling can be framed in policy debates as frustrating free Uncertainty also exists on the extent that government product- trade, economic opportunity, and the sovereignty of the producing labeling programs, as opposed to other types of public regulatory mea- (often developing) country,29 or as promoting the freedom of individ- sures, are vulnerable to trade challenges (United States—Tuna/Dolphin uals in the consuming (often developed) country to have access to in- III Panel Report). The TBT Agreement explicitly distinguishes between formation that will enable them to express preferences for reducing technical regulations (with which compliance is mandatory) and techni- the likelihood of climate change through less carbon-intensive con- cal standards (which are voluntary), but both are potentially subject to sumption (Vandenbergh and Cohen, 2010). In the absence of trade dis- challenge. Programs are treated as mandatory if the label is a legally bind- putes over carbon labeling systems, it is too early to tell how these ing market-access requirement (Tietje, 1995), and an expansive view has issues will arise and whether they will affect the treatment of public been taken about when a measure is mandatory (United States—Tuna/ and private carbon labeling systems by the trade regime. Dolphin III Panel Report).24 Government consumer product labels are more likely to be viewed as WTO-compliant when they are voluntary, market-based, and transparent,25 but the recent Tuna/Dolphin decisions

26 See, e.g., United States—Hot-Rolled Steel, paragraph 81, which concludes that “[i]n 22 For example, the 2011 Tuna/Dolphin panel found the United States not in violation principle, any act or omission attributable to a WTO Member can be a measure of that of TBT 2.1, and in violation of TBT 2.2, but declined to rule on the GATT claims (United Member for purposes of dispute settlement proceedings.” States—Tuna/Dolphin III, Panel Report). More recently, the Appellate Body found the 27 Actions “taken by private parties does not rule out the possibility that it may be United States in violation of TBT 2.1 but not TBT 2.2, but again declined to rule on deemed to be governmental if there is sufﬁcient government involvement with it. It the GATT claims. The Appellate Body concluded, however, that the panel should have is difﬁcult to establish bright-line rules in this regard, however. Thus, that possibility conducted a GATT analysis in case the Appellate Body concluded that a TBT violation will need to be examined on a case-by-case basis” (Japan—Film). had not occurred (United States—Tuna/Dolphin III, AB Report). 28 The extent to which government procurement policies are subject to review under 23 For example, U.S. dolphin safe tuna labeling was determined to be more trade re- the TBT, Article 1.4, is unclear, as is the extent to which procurement policies might af- strictive than necessary because the “labeling provisions only partly address the legit- fect whether a dispute panel would take jurisdiction over a private carbon label. Pro- imate objectives pursued,” and Mexico, the complaining Member, provided the WTO curement policies are reviewed under the Agreement on Government Procurement, dispute panel with a less trade-restrictive alternative capable of achieving the same which is one of the “plurilateral agreements” within the WTO structure. As such, the level of protection as the objective pursued by the original measure (United States— GPA is binding only on those WTO members who choose to adhere to it, unlike the Tuna/Dolphin II). GATT and TBT, which are binding upon all WTO members. 24 For example, the 2011 Tuna/Dolphin WTO panel decision on Mexico's challenge 29 See, e.g., United States—Gasoline, which concludes that “WTO Members have a was to a U.S. law that did not require labeling of tuna as dolphin-safe but set standards large measure of autonomy to determine their own policies on the environment (in- that had to be met if a label was used (United States—Tuna/Dolphin III Panel Report; cluding its relationship with trade), their environmental objectives and the environ- United States—Tuna/Dolphin III AB Report). mental legislation they enact and implement. So far as concerns the WTO, that 25 See WTO Committee on Trade and the Environment, http://www.wto.org/english/ autonomy is circumscribed only by the need to respect the requirements of the General tratop_e/envir_e/labelling_e.htm. Agreement and the other covered agreements.” M.A. Cohen, M.P. Vandenbergh / Energy Economics 34 (2012) S53–S63 S61

5.3. Product label design issues substitutes, and many goods have complex or variable supply chains that will require high transaction costs to assess.31 All these concerns Although carbon labels are relatively new, product labels have a long are sound, and a carbon-labeling system, particularly a private one, is un- history dating back to the 1960s in the United States. Both government- likely to be widely adopted and achieve substantial carbon emissions re- mandated and voluntary third-party certified labels now exist for a host ductions unless a subset of the most promising goods can be identified. of products and attributes ranging from cigarettes, nutritional content of The challenge is to identify those goods that will yield significant food, organic produce, flammable materials, and hazardous chemicals. emission reductions at comparatively lower cost than other near-term Beginning with Viscusi and Magat (1987),manystudieshavefocused options. A carbon-labeling system should be compared not with ideal on consumer responses to labels with the goal of designing meaningful alternative instruments but with the viable options for the relevant communication vehicles that will not mislead and will provide action- time period.32 With the current international deadlock, a carbon tax able information. A recent paper by Cohen and Viscusi (forthcoming) or cap-and-trade system offers only limited near-term competition to summarizes this literature and makes recommendations related to car- a carbon-labeling regime. Meanwhile, a carbon-labeling system may bon footprint labels. Many of these recommendations involve the design be able to circumvent collective action problems these international of labels to ensure that consumers “receive, process, believe and use” the and domestic public measures face, generate near-term emissions reduc- information on them. tions to buy time for more comprehensive remedies, build public support One critical issue that arises in the context of our analysis is how for the more comprehensive remedies, and provide supply chain infor- to ensure that the product label not only conveys static information mation that will facilitate the implementation of anti-leakage provisions about the product life cycle but also provides actionable information inthemorecomprehensiveremedies(Vandenbergh et al., 2011). For ex- to consumers when their product use decisions have a significant im- ample, a successful U.S. cap-and-tradeprogramwillrequiresomeform pact on carbon emissions. For example, life-cycle assessments have of border allowance system, which presumes the ability to gather and determined that water temperature is the most significant factor in analyze many of the types of carbon emission data for consumer goods carbon emissions resulting from home laundry activities (White, that will be developed through carbon labeling. 2009), yet consumers often use warmer water than called for on the Here, we begin to sketch out principles for determining which product instructions. Thus, in attempting to reduce their product goods are best suited for carbon labels. The first is that the screening life-cycle carbon emissions, Proctor & Gamble focused their attention methodology must identify goods for which changes in consumption on an education campaign to convince U.K. consumers to use cold (whether substitution or reduced use) could yield relatively large car- water. This example raises two interesting questions from a carbon- bon emission reductions. Although this is a commonsense concept on labeling perspective. First, should the carbon footprint be calculated the surface, it is important not to overlook the fact that not all prod- based on actual consumer use or on the instructions that call for cold ucts have sufficiently large carbon footprints to justify labeling efforts water? Note that Publicly Available Specification 2050 resolves this designed to induce substitution with other goods or less consumption issue by calling for “actual usage” to be the guiding principle (British of the labeled good. Standards Institution, 2011, Section 6.4.9.2). While this is probably ap- A second principle is that the screening methodology must be able propriate for carbon label standards, it is also a rather static view and to account for the costs of information gathering. As discussed above, it would seem preferable to find ways to educate consumers that carbon the complexity of the issues at each of the important stages in the life emissions would be significantly lower if they used the product as cycle of a product suggests that a system that seeks a high degree of pre- intended and also to periodically update the label to reflect any changes cision will collapse under the weight of heavy transaction costs. Goods in actual usage. Thus, the second issue is whether the label should be with more complex or shifting supply chains, for example, may not be used to help educate consumers about the proper product use and to promising initial candidates for carbon labeling. For many goods, how- provide firms with the incentive to change consumers' “actual usage” ever, the error bands may not overlap: limited information may result over time? Although some exceptions might apply, label clutter is likely in a wide range of potential emissions that are attributable to the to make this approach less than desirable. Complementary approaches good, but even the high-end carbon emission estimate of a substitute might include point-of-sale brochures, product inserts, mobile phone good is unlikely to be as high as the low end of the other (e.g., chicken apps, as well as other marketing campaigns as evidenced by P&G's at- versus beef). tempt to convince consumers to use cold water. A third principle is that the screening for the most promising An important caveat to our analysis of carbon labels is that many products should account for each step in the life cycle of a good, in- of the existing product labels are based on multiattribute criteria, cluding production, transport, storage and sales, consumption, and and terms such as “eco-labels” or “sustainability” are used to convey disposal. However, it may not be necessary to conduct a detailed anal- their benefits. For example, the Green Seal certification process30 in- ysis of all aspects of any one step in the life cycle of a good. For exam- cludes product ingredients (toxics, carcinogens, etc.), worker safety, ple, it may be possible to screen for transport emissions for some human rights, water pollution, and greenhouse gas emissions (among goods based on the distance traveled. For other goods, the carbon in- other issues). A separate carbon footprint label might confuse consumers tensity of the means of transport may account for a sufficiently large and/or crowd out labels that focus on other social issues. While this is not share of total emissions to serve as a screening criterion. For some ag- aninevitableoutcome,caremustbetakennottoimposeexcessivecosts ricultural products, seasonality may have such a large effect on the on other product attribute labels that serve important social roles. carbon footprint of a category of goods that screening criteria can be developed to identify categories of the most promising goods. One example is field-grown tomatoes in the summer versus hothouse toma- 5.4. Which products should be labeled? toes in the winter in England. A fourth principle is that the screening methodology should Labeling the carbon emissions associated with millions of consumer account for the behavioral plasticity of consumers (Dietz et al., goods would be expensive, and the marginal costs of labeling will ex- 2009)—i.e. the extent to which behavior is likely to change following ceed the benefits for many goods. Critics of carbon labeling have noted that labeling in itself is not a comprehensive remedy for carbon 31 emissions, some goods within a product category have only minor dif- For example, see “Carbon Footprints: Following the Footprint”, Economist, June 2, ferences in carbon emissions, other goods do not have low-carbon 2011. 32 Dan Farber (2011) calls the tendency in law and economics to dismiss good proposals based on the existence of conceivable but non-viable alternatives the “fallacy 30 See www.greenseal.org. of hypothetical alternatives.” S62 M.A. Cohen, M.P. Vandenbergh / Energy Economics 34 (2012) S53–S63 a new policy (in this case, introduction of a carbon label), whether the Acknowledgments behavior change involves reduced consumption of the good or substitution with a good with a smaller carbon footprint.33 Goods differ both in We gratefully acknowledge the research assistance of Madeline the extent to which labels are likely to affect consumer purchasing de- Gottlieb and Marcy Nicks Moody. Helpful comments were received cisions (Golden, 2010, 11)34 and the substitutes that consumers are from Jeffrey Dunoff, Will Martin, Sharon Shewmake and two anony- likely to select when changing consumption patterns. The elasticity of mous referees. All remaining errors are those of the authors. demand and carbon footprints of likely substitutes are known for many consumer goods (e.g., Okrent and Alston, 2011) and can be estimated for others. What is perhaps less well known is the perceived sub- References stitutability of carbon footprint for other product attributes. Similar to Andreoni, J., 1990. Impure altruism and donations to public goods: a theory of warm- the impact of a change in price, new information about a product attri- glow giving. Econ. J. 100, 464–477. bute (carbon) might induce consumers to substitute into other prod- Baksi, S., Bose, P., 2007. Credence goods, efficient labeling policies, and regulatory en- – ucts (or reduce consumption altogether). forcement. Environ. Resour. Econ. 37 (2), 411 430. fi Baron, D.P., Diermeier, D., 2007. Strategic activism and nonmarket strategy. J. Econ. For a voluntary government or private system, an important fth Manag. Strateg. 16 (3), 599–634. principle is that the screening of goods for labeling should account for Bird, L., Sumner, J., 2010. Green power marketing in the United States: a status report. those goods that firms may have the greatest incentive to label (i.e., the Technical Report NREL/TP-6A20-4903. National Renewable Energy Laboratory, fi Washington, DC. http://www.nrel.gov/docs/fy11osti/49403.pdf. behavioral plasticity of the rm). For instance, if a good will be subject Bjørner, T.B., Lars Gårn, L.H., Russell, C.S., 2004. Environmental labeling and consumers' to reputation campaigns or boycotts, the firm selling it has an incentive choice—an empirical analysis of the effect of the Nordic Swan. J. Environ. Econ. to participate in a labeling system (Lenox and Eesley, 2009). Similarly, a Manag. 47 (3), 411–434. Borchers, A.M., Duke, J.M., Parsons, G.R., 2007. Does willingness to pay for green energy good is amenable to labeling if carbon-friendly changes to the supply differ by source? Energy Policy 35 (6), 3327–3334. chain or substitution opportunities among the firm's goods will lead to British Standards Institution, 2011. PAS 2050: 2011—Specification for the Assessment energy cost savings (Carbon Trust, 2006 11, 13; Humes, 2011). of the Life Cycle Greenhouse Gas Emissions of Goods and Services. London http:// www.bsigroup.com/upload/Standards%20&%20Publications/Energy/PAS2050.pdf. Brown, R., Webber, C., Koomey, J., 2002. Status and future directions of the ENERGY 6. Concluding remarks STAR program. Energy 27 (5), 505–520. Carbon Trust, 2006. Carbon Footprints in the Supply Chain: The Next Stop for Business . (Publication ID CTC616. London). Carbon footprint product labeling is in its infancy as companies, Charnovitz, S., 2002. The law of environmental “PPMs” in the WTO: debunking the third-party certification organizations, and government agencies ex- myth of illegality. Yale J. Int. Law 27 (1), 59–110. Clark, C.F., Kotchen, M.J., Moore, M.R., 2003. Internal and external influences on pro- periment with methodology and label design. While we are already environmental behavior: participation in a green electricity program. J. Environ. beginning to see an emerging global standardized protocol for esti- Psychol. 23 (3), 237–246. mating the life-cycle carbon footprint of products, no such standard- Cohen, M.A., Viscusi, W.K., forthcoming. The role of information disclosure in climate mitigation policy. Clim. Chang. Econ. ized approach to labeling has yet emerged. Darby, M.R., Karni, E., 1973. Free competition and the optimal amount of fraud. J. Law The theory behind carbon footprint labeling is clear: without infor- Econ. 16 (1), 67–68. mation about the greenhouse gas implications of product choices, con- Davis, S.J., Caldeira, K., 2010. Consumption-based accounting of CO2 emissions. Proc. – sumers are unable to make informed choices about which products to Natl. Acad. Sci. 107 (12), 5687 5693. Dietz, T., Gardner, G.T., Gilligan, J., Stern, P.C., Vandenbergh, M.P., 2009. Household ac- purchase and how to use them. However, the theory is also clear that tion can provide a behavioral wedge to rapidly reduce U.S. carbon emissions. Proc. without credible third-party certification, it is unlikely that a meaningful Natl. Acad. Sci. 106 (44), 18452–18456. Dranove, D., Jin, G.Z., 2010. Quality disclosure and certification: theory and practice. carbon-labeling market will develop. Moreover, because climate change – fi J. Econ. Lit. 48 (4), 935 963. is a global problem and international trade accounts for a signi cant por- Dunoff, J., 1994. Institutional misfits: the GATT, the ICJ & trade–environment disputes. tion of carbon emissions, any meaningful carbon-labeling system will re- Mich. J. Int. Law 15, 1043–1128. quire a globally accepted, uniform methodology for calculating life-cycle Earley, R., Kang, L., An, F., Green-Weiskel, L., 2011. Electric vehicles in the context of sustainable development in China. Commission on Sustainable Development. emissions and labeling products. Background Paper No. 9. CSD19/2011/BP9. United Nations Department of Econom- The empirical evidence on the potential impact of carbon labeling ic and Social Affairs, New York. http://www.un.org/esa/dsd/resources/res_pdfs/ is sparse but growing. Some product segments have a large potential csd-19/Background-Paper-9-China.pdf. fi Energy Information Administration, 2008. Emissions of Greenhouse Gases in the Unit- for carbon emission reductions, and signi cant consumer segments ed States 2007. U.S. Department of Energy, Washington, DC. care about carbon emissions. The latter are likely to grow as more Farber, D.A., 2011. Climate Justice. (July 10, Available at SSRN:) http://ssrn.com/ information is disclosed and as life-cycle analyses uncover more abstract=1883186. — Fischer, C., Fox, A.K., 2011. The role of trade and competitiveness measures in US cli- ways to reduce carbon emissions without increasing or even while mate policy. Am. Econ. Rev. Pap. Proc. 101 (3), 258–262. decreasing—the cost to consumers. Perhaps as important, we expect Gardner, G.T., Stern, P.C., 2008. The short list: the most effective actions U.S. households carbon labeling to have a role in increased supply chain pressures can take to curb climate change. Environment 50 (5), 12–23. by manufacturers and large chain stores to reduce the carbon emis- Ghertner, D.A., Fripp, M., 2007. Trading away damage: quantifying environmental leakage through consumption-based, life-cycle analysis. Ecol. Econ. 63 (2–3), sions of products they carry. 563–577. While the methodological and legal challenges to the establish- Golden, J.S. (Ed.), 2010. An Overview of Ecolabels and Sustainability Certifications in ment of effective carbon-labeling programs are significant, careful the Global Marketplace. Interim Report Document #1010-10-1. Nicholas Institute, Duke University, Durham, NC. http://center.sustainability.duke.edu/sites/default/ analysis and selection of product categories to be labeled could ulti- files/documents/ecolabelsreport.pdf. mately bring about significant reductions in carbon emissions in a Greenhouse Gas Protocol, 2011. Product Life Cycle Accounting and Reporting Standard. manner that is cost-effective and consistent with international trade World Resources Institute and World Business Council for Sustainable Develop- ment, Washington, DC . http://www.ghgprotocol.org/standards/product-standard. standards. Greening, L., Greene, D., Difiglio, C., 2000. Energy efficiency and consumption – the rebound effect – a survey. Energy Policy 28, 389–401. 33 York et al. (2002) proposed the term “behavioral plasticity” to describe this phe- Grossman, S., 1981. The informational role of warranties and private disclosure about product quality. J. Law Econ. 24 (3), 461–489. nomenon in the household behavior context. Heal, G., Kunreuther, H., 2010. Social reinforcement: cascades, entrapment, and tip- 34 Golden notes, “Ecolabels do not impact purchasing decisions equally across prod- ping. Am. Econ. J.: Microeconomics 2 (1), 86–99. uct categories. Four variables in particular determine consumption practices when it Humes, E., 2011. Force of Nature: The Unlikely Story of Wal-Mart's Green Revolution. comes to buying green: purchase visibility, consumption visibility, durability, and HarperCollins, New York. perishability. Ecolabels matter more for nondurable, frequently used, and highly vis- Ibanez, L., Grolleau, G., 2008. Can ecolabeling schemes preserve the environment? ible consumer goods. So, being an environmentally responsible soft drink producer Environ. Resour. Econ. 40 (2), 233–249. should carry more of a competitive advantage than being an ecofriendly insulation Ippolito, P.M., Mathios, A.D., 1990. The regulation of science-based claims in advertis- manufacturer.” ing. J. Consum. Policy 13 (4), 413–445. M.A. Cohen, M.P. Vandenbergh / Energy Economics 34 (2012) S53–S63 S63

Jacobsen, G., Kotchen, M.J., Vandenbergh, M.P., 2012. The behavioral response to volun- Vandenbergh, M.P., Dietz, T., Stern, P.C., 2011. Time to try carbon labelling. Nat. Rep. tary provision of an environmental public good: evidence from residential electric- Clim. Chang. 1, 4–6. ity demand. Eur. Econ. Rev. http://dx.doi.org/10.1016/j.euroecorev.2012.02.008. Viscusi, W.K., Magat, W.A., 1987. Learning about Risk: Consumer and Worker Re- Joshi, M., 2004. Are eco-labels consistent with World Trade Organization agreements? sponses to Hazard Information. Harvard University Press, Cambridge. J. World Trade 38 (1), 69–92. Vranes, E., 2011. Climate labelling and the WTO: the 2010 EU ecolabelling programme Kals, E., Schumacher, D., Montada, L., 1999. Emotional affinity toward nature as a mo- as a test case under WTO law. In: Hermann, C., Terhecte, J.P. (Eds.), European Year- tivational basis to protect nature. Environ. Behav. 31 (2), 178–202. book of International Economic Law 2011, vol. 2. Springer, Amsterdam, The Neth- Koning, A., Schowanek, D., Dewaele, J., Weisbrod, A., Guinee, J., 2009. Uncertainties in a erlands, pp. 205–237. carbon footprint model for detergents; quantifying the confidence in a compara- Watson, D., Moll, S., 2008. Environmental benefits and disadvantages of economic spe- tive result. Int. J. Life Cycle Assess. http://dx.doi.org/10.1007/s11367-009-0123-3. cialisation within global markets, and implications for SCP monitoring. In: Ken, Kotchen, M.J., 2005. Impure public goods and the comparative statics of environmen- T.G., Tukker, A., Vezzoli, C., Ceschin, F. (Eds.), Proceedings: Refereed Sessions I–II, tally friendly consumption. J. Environ. Econ. Manag. 49 (2), 281–300. Sustainable Consumption and Production: Framework for Action. TNO Built Envi- Kotchen, M.J., 2006. Green markets and private provision of public goods. J. Polit. Econ. ronment and Geosciences, Delft: The Netherlands. 114 (4), 816–834. White, P., 2009. Building a sustainability strategy into the business. Corp. Gov. 9 (4), Kotchen, M.J., Moore, M.R., 2007. Private provision of environmental public goods: 386–394. household participation in green-electricity programs. J. Environ. Econ. Manag. Wiedmann, T., Wood, R., Lenzen, M., Minx, J., Guan, D., Barrett, J., 2008. Development of 53 (1), 1–16. an Embedded Carbon Emissions Calculator: Producing a Time Series of Input– Kysar, D., 2004. Preferences for processes: the process/product distinction and the reg- output Tables and Embedded Carbon Dioxide Emissions for the UK by Using a ulation of consumer choice. Harv. Law Rev. 118 (2), 525–642. MRIO Data Optimisation System. Department for Environment, Food and Rural Lenox, M.J., Eesley, C.E., 2009. Private environmental activism and the selection and re- Affairs, London. sponse of firm targets. J. Econ. Manag. Strateg. 18 (1), 45–73. Wiser, R., Bolinger, M., Holt, E., 2000. Customer choice and green power marketing: a Liebenstein, H., 1950. Bandwagon, Snob and Veblen effects in the theory of consumer's critical review and analysis of experience to date. Proceedings of the 2000 ACEE's demand. Q. J. Econ. 64 (2), 183–207. Summer Study on Energy Efficiency in Buildings (Available at http://eetd.lbl.gov/ Lyon, T.P., Maxwell, J.W., 2011. Greenwash: corporate environmental disclosure under EA/EMP/). threat of audit. J. Econ. Manag. Strateg. 20 (1), 3–41. York, R., Rosa, E., Dietz, T., 2002. Bridging environmental science with environmental Maiteny, P.T., 2002. Mind in the gap: summary of research exploring “inner” influences policy: plasticity of population, affluence and technology. Soc. Sci. Q. 83 (2), 18–34. on pro-sustainability learning and behaviour. Environ. Educ. Res. 8 (3), 299–306. Michaud, C., Llerena, D., Joly, I., forthcoming. Willingness to pay for environmental attributes of non-food agricultural products: a real choice experiment. Eur. Rev. LEGAL REFERENCES Agric. Econ. http://dx.doi.org/10.1093/erae/jbs025. — — Mintel Group, 2011. Green Marketing U.S. April 2011. Agreement on Government Procurement, 1994. Marrakesh Agreement Establishing the – Nelson, P., 1970. Information and consumer behaviour. J. Polit. Econ. 78 (2), 311 329. World Trade Organizations, Annex 4(b). (April 15, 33 I.L.M. 1125). Okrent, A.M., Alston, J.M., 2011. The demand for food in the United States: a review of Argentina—Leather, 2000. Panel Report, Argentina—Measures Affecting the Export of the literature, evaluation of previous estimates, and presentation of new estimates Bovine Hides and the Import of Finished Leather, WT/DS155/R . (December 19). of demand. Monograph, 48. Giannini Foundation of Agricultural Economics, Berke- Brazil—Tyres, 2007. Appellate Body Report, Brazil—Measures Affecting Imports of Re- ley, CA. http://giannini.ucop.edu/Monographs/48-FoodDemand.pdf. treaded Tyres, WT/DS332/AB/R . (December 3). Ottman, J., 2011. The New Rules of Green Marketing. Barrett-Koehler, San Francisco. Canada—Dairy, 1999. Panel Report, Canada—Measures Affecting the Importation of Pauwelyn, J., 2004. The puzzle of WTO safeguards and regional trade agreements. J. Int. Milk and the Exportation of Dairy Products, WT/DS103/R, WT/DS113/R . (May 17). – Econ. Law 7 (1), 109 142. Convention on International Trade in Endangered Species of Wild Fauna and Flora, – Peters, G.P., Hertwich, E.G., 2008. Trading Kyoto. Nat. Rep. Clim. Chang. 2, 40 41 http:// March 3, 1973, 27 U.S.T. 1087, T.I-A.S. No. 8249, 993 U.N.T.S. 243. dx.doi.org/10.1038/climate.2008.25. EC–Asbestos, 2001. Appellate Body Report, European Communities—Measures Affect- Peters, et al., 2011. Growth in emission transfers via international trade from 1990 to ing Asbestos and Asbestos-containing Products, WT/DS135/AB/R . (March 12). – 2008. Proc. Natl. Acad. Sci. 108 (21), 8903 8908. EC—Sardines, 2002. Appellate Body Report, European Communities—Trade Description Roberts, T., 2011. Innovation as governance: a functional typology of private gover- of Sardines, WT/DS231/AB/R . (September 26). – nance institutions. Duke J. Environ. Law Policy 22 (1), 67 144. General Agreement on Tariffs and Trade, October 30, 1947, 61 Stat. A-11, 55 U.N.T.S. Sorrell, S., Dimitropoulos, J., Sommerville, M., 2009. Empirical estimates of the direct 194. – rebound effect: a review. Energy Policy 37, 1356 1371. Japan—Film. Panel Report, Japan—Measures Affecting Consumer Photographic Film and Stavins, R.N., 2008. A meaningful U.S. cap-and-trade system to address climate change. Paper, ¶ 10.56, WT/DS44/R (March 31, 1998).TBT Agreement (Agreement on Tech- – Harv. Environ. Law Rev. 32 (2), 293 371. nical Barriers to Trade), April 15, 1994, Marrakesh Agreement Establishing the Swallow, S.K., Sedjo, R.A., 2000. Eco-labeling consequences in general equilibrium: a World Trade Organization, Annex 1A, 1868 U.N.T.S. 120. – graphical assessment. Land Econ. 76 (1), 28 36. United States—Gasoline, 1996. Appellate Body Report, United States—Standards for Teisl, M.F., Hicks, R.L., 2002. Can eco-labels tune a market? Evidence from dolphin-safe Reformulated and Conventional Gasoline, 29–30, WT/DS2/AB/R . (April 22). – labeling. J. Environ. Econ. Manag. 43 (3), 339 359. United States—Hot-Rolled Steel, 2003. Appellate Body Report, United States—Sunset Thøgersen, J., Ölander, F., 2003. Spillover of environment-friendly consumer behaviour. Review Of Anti-Dumping Duties On Corrosion-Resistant Carbon Steel Flat Products – J. Environ. Psychol. 23 (3), 225 236. From Japan, Wt/Ds244/Ab/R . (December 15). Tietje, C., 1995. Voluntary eco-labelling programmes and questions of state responsi- United States—Shrimp/Turtle I, 1998. Appellate Body Report, United States—Import – bility in the WTO/GATT legal system. J World Trade 29 (5), 123 158. Prohibition of Certain Shrimp and Shrimp Products, WT/DS58/AB/R . (October 12). fi U.S. EPA (Environmental Protection Agency), Of ce of Inspector General, 2008. Im- United States—Shrimp/Turtle II, 2001. Appellate body report, United States—import fi provements Needed to Validate Reported ENERGY STAR Bene ts. (Report No. 09- prohibition of certain shrimp and shrimp products, recourse to Article 21.5 of the P-0061. Washington, DC) http://www.epa.gov/oig/reports/2009/20081217-09-P- DSU by Malaysia. WT/DS58/AB/RW. (October 22). 0061.pdf. United States—Tuna/Dolphin I, 1991. Panel Report, United States—Restrictions On Im- Vanclay, J.K., Shortiss, J., Auselbrook, S., Gillespie, A.M., Howell, B.C., Johanni, R., Maher, ports Of Tuna. Gatt B.I.S.D . (September 3, 39th Supp.). M.J., Mitchell, K.M., Stewart, M.D., Yates, J., 2011. Customer response to carbon la- United States—Tuna/Dolphin III AB Report, 2012. Appellate body report, United States— – belling of groceries. J. Consum. Policy 34, 153 160. measures concerning the importation, marketing and sale of tuna and tuna products. Vandenbergh, M.P., 2007. The new Wal-Mart effect: the role of private contracting in WT/DS381/AB/R. (May 16). – global governance. U.C.L.A. Law Rev. 54, 914 970. United States—Tuna/Dolphin III Panel Report, 2011. Panel report, United States— Vandenbergh, M.P., Cohen, M.A., 2010. Climate change governance: boundaries and measures concerning the importation, marketing and sale of tuna and tuna prod- – leakage. NYU Environ. Law J. 18, 221 292. ucts. WT/DS381/R. (September 15). Vandenbergh, M.P., Steinemann, A., 2007. The carbon-neutral individual. NYU Law Rev. 82 (6), 1673–1745.